There is a fundamental gap in understanding the neuro-behavioral time-course of reading recovery following stroke. It is not known whether recovery of reading is associated with improved blood circulation and neural activity of the peri-infarct area, or whether circulation in the unaffected, contralateral brain areas contributes to functional recovery. Continued existence of this gap represents an important problem because, until it is filled, knowledge about optimal timing and targets for restorative therapies to improve reading will remain out of reach. The long-term goal is to develop clinical prognostic and therapeutic tools to improve reading limitations. The overall objective of this project is to characterize the neural mechanisms of recovery from stroke-induced reading impairments. The central hypothesis of this study is that subacute-to-chronic recovery of reading ability is potentiated by reperfusion (improved blood flow) of the left reading network. Reperfusion promotes the return of neural activity and supports behavioral recovery. In contrast, increased perfusion of the right brain areas represents a maladaptive response and is associated with worse chronic reading ability. This hypothesis has been formulated on the basis of preliminary data. The rationale for the proposed research is that once the neural recovery from stroke-induced reading impairments is characterized, clinicians will be able to predict individual recovery potential and select appropriate rehabilitation goals. Guided by strong preliminary data, this hypothesis will be tested by pursuing two specific aims: 1) Determine whether post-stroke changes in cerebral blood flow predict recovery of reading ability; and 2) Determine whether increased baseline perfusion of the left reading network is paralleled by greater reading-related neural activation. Under the first aim, perfusion in the peri-infarct tissue, left reading network, and homologous right brain areas will be examined among left stroke patients with reading impairments. A non-invasive measure of perfusion (Arterial Spin Labeling, ASL MRI) will be applied longitudinally: <4 weeks post-stroke (sub-acute) and >3 months post-stroke (chronic) to test if increased perfusion of the left reading network is coupled with an improvement of reading accuracy. Under the second aim, reading-induced brain activity will be recorded in the same group of patients using perfusion- based functional MRI. The effect of task-related neural activity on recovery will be separated from the effect of cerebral blood flow by statistically controlling for baseline circulation and by modeling independent contributions of perfusion and BOLD (Blood Oxygen Level Dependent) signals in order to estimate cerebral oxygen consumption rate, which is directly related to neural activity. The conceptual innovation of this study is that it offers the opportunity to examine reading recovery as distinct from other functions. The methodological innovation is in using simultaneously-acquired perfusion and BOLD signals, helping to measure longitudinal changes in circulation and neural activity with unprecedented precision. The proposed research is significant because it can help greatly improve the effectiveness of post-stroke rehabilitation.